1. Field of the Invention
The invention relates to a method, system and computer program product for grading asset-backed securities including collateralized mortgage obligations by assigning a relative grade, including a prepayment risk component and a complexity component, which may take into account one or more of weighted average life, volatility of weighted average life at moderate and stressed yield curve shifts, and complexity, in the absence of price information.
2. Description of the Related Art
Asset-backed securities are an important financing technique that permits illiquid assets such as home mortgages, credit card debt or automobile loans to be pooled in an economical and practical manner. The underlying assets are pooled to secure debt, and the resulting pools may be bought and sold among investors and financial institutions. The principal and interest streams associated with the pools can also be bought and sold, and transferred to other entities. Collateralized mortgage obligations (CMOs) are mortgage-backed securities which allow cash flows to be directed so that different classes of securities with different maturities and coupons can be created. These classes of securities are also referred to as bonds or tranches.
Mortgage-backed securities differ from other asset-backed securities in two important respects. First, most mortgages are subject to prepayment at the option of the obligor, so there is a risk that a bond may not be outstanding as long as an investor anticipates due to prepayment of the mortgages that underlie the mortgage-backed security. Second, the majority of mortgage-backed securities have the advantage of reduced credit risk through guarantees provided by the U.S. government (through Ginnie Mae) or by government sponsored enterprises such as Fannie Mae and Freddie Mac (“GSEs”).
In comparison to other asset-backed securities, mortgage-backed securities have disadvantages, such as substantially longer maturities. For example, a pool of thirty year mortgages has a much longer maturity than a pool of five year automobile loans.
As is the case with almost all asset-backed securities, mortgage-backed securities such as CMOs are sensitive to interest rate changes. The interest rate risk associated with CMOs is reflected in one measure by prepayment speed. As interest rates fall, it becomes more likely that an individual mortgagor will refinance. Refinancing cuts off the payment stream that was associated with the original mortgage. The resulting interruption of an investor's income stream expectations significantly affects the value of the CMO.
It is therefore important to quantify and distinguish prepayment risk as a separate component of the quality and value of any asset-backed security, especially a CMO. The financial markets have developed methods for quantifying prepayment risk and prepayment expectations in CMOs. Existing grading systems may be based upon empirical data. Conventional, commercially available or proprietary grading systems, such as those available through Bloomberg (e.g., GRADE, C.FLUX and FFIEC), may rely on a price input, i.e., the price an investor is willing to pay for the CMO. While price may be an appropriate input for a market-based measure of a CMO's sensitivity to changes in the market (e.g., interest rate changes), market prices are not readily available for many CMOs because, in many cases, they are not widely traded. Moreover, the price of a CMO may be affected by the liquidity of the underlying assets or the marketing effectiveness of the investor or financial institution holding the security.
Prepayment assumptions for some widely traded securities are published by financial reporting services. The published prepayment assumptions are based upon market consensus and are derived from participants in the securities markets such as traders, underwriters and the financial institutions issuing the securities. The published prepayment assumptions may be based at least in part upon a comparison of current and past pricing.
Existing methods for determining prepayment risk are deficient in the richness and quality of the information they provide. Weighted average life (WAL) is one conventional measure of prepayment risk. A WAL value provides some indication to the investor of the period of maturity of the underlying security (i.e., the average time that each principal dollar in a pool is expected to be outstanding). Importantly, however, WAL does not provide investors with a measure of the sensitivity of the underlying security to changes in the market, for example, sensitivity to interest rate changes.
Likewise, some conventional methods for grading asset-backed securities such as CMOs rely on proprietary methodology for which the calculations and/or mechanics are not accessible or evident to the investor. Furthermore, such proprietary methodology may be unable to characterize a CMO with enough information for an investor to make an investment decision. These proprietary methodologies may also be based in part on difficult-to-obtain price information.
The proprietary grading systems mentioned above may be useful for bringing liquidity to the market, however proprietary systems need price information from which to derive a security's value and/or quality. True price information is difficult to obtain using proprietary systems and, for a particular security, may be determinable only upon trading of the underlying security (e.g., buying or selling the security to determine its market price). This is especially problematic for securities that are thinly traded (e.g., traded in low volume or traded on a secondary market). In the absence of richer and more detailed information regarding the pricing and/or quality of securities, investors may be faced with choosing between two CMOs with apparently equivalent grades or prepayment risks; however, the underlying assets may be substantially different with regard to their sensitivity to interest rate change and/or other changes in the market. The end effect of conventional and proprietary grading processes is to limit investor participation. Thus, the lack of grading systems that do not rely on price information makes the markets for CMOs less liquid and provide a less attractive investment alternative.